This study was undertaken to assess the clinical significance of changes in muscle glycogen in the calculation of the energy requirements of patients fed intravenously. The glycogen content of the vastus or rectus muscle of 27 ill surgical patients with serious complications of the gastrointestinal tract was determined before and after a course of intravenous nutrition with various quantities of glucose as the caloric source. The quantity of glycogen in muscle before intravenous nutrition for all patients was 37 +/- 23 mg/g dry muscle; this was increased significantly (p less than .01) to 69 +/- 44 mg/g dry muscle after 14 days of feeding. In 5 patients who received intravenous nutrition for another 2 weeks, the glycogen content of the vastus muscle increased from a mean value of 80 mg/g dry muscle on day 14 of feeding to 106 mg/g on day 28, but this was not significant. The administration of glucose is invariably associated with an increase in the stores of muscle glycogen but the calorie equivalent of this gain in glycogen is small and, by our calculations, is no more than 530 kcal over the 14 days. We suggest that this gain is small relative to the energy expenditure of this type of patient and is of little clinical significance in the determination of their calorie requirements.